Biochemistry 2010, 49, 9911–9921 9911 DOI: 10.1021/bi100974v
)
Adaptation to a High-Tungsten Environment: Pyrobaculum aerophilum Contains an Active Tungsten Nitrate Reductase† Simon de Vries,‡ Milica Momcilovic,§ Marc J. F. Strampraad,‡ Julian P. Whitelegge, Ashkan Baghai,^ and Imke Schr€oder*,^
Laboratory of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands, § Biomedical Engineering Interdepartmental Program, Department of Chemistry and Biochemistry, and ^ Department of Microbiology, Immunology, and Molecular Genetics, University of California at Los Angeles, California 90095, United States )
‡
Received June 16, 2010; Revised Manuscript Received September 3, 2010 ABSTRACT: Nitrate reductases (Nars) belong to the DMSO reductase family of molybdoenzymes. The hyperthermophilic denitrifying archaeon Pyrobaculum aerophilum exhibits nitrate reductase (Nar) activity even at WO42- concentrations that are inhibitory to bacterial Nars. In this report, we establish that the enzyme purified from cells grown with 4.5 μM WO42- contains W as the metal cofactor but is otherwise identical to the Mo-Nar previously purified from P. aerophilum grown at low WO42- concentrations. W is coordinated by a bis-molybdopterin guanine dinucleotide cofactor. The W-Nar has a 2-fold lower turnover number (633 s-1) but the same Km value for nitrate (56 μM) as the Mo-Nar. Quinol reduction and nitrate oxidation experiments monitored by EPR with both pure W-Nar and mixed W- and Mo-Nar preparations suggest a monodentate ligation by the conserved Asp241 for W(V), while Asp241 acts as a bidentate ligand for Mo(V). Redox titrations of the Mo-Nar revealed a midpoint potential of 88 mV for Mo(V/IV). The Em for W(V/IV) of the purified W-Nar was estimated to be -8 mV. This relatively small difference in midpoint potential is consistent with comparable enzyme activities of W- and Mo-Nars. Unlike bacterial Nars, the P. aerophilum Nar contains a unique membrane anchor, NarM, with a single heme of the oP type (Em = 126 mV). In contrast to bacterial Nars, the P. aerophilum Nar faces the cell’s exterior and, hence, does not contribute to the proton motive force. Formate is used as a physiological electron donor. This is the first description of an active W-containing Nar demonstrating the unique ability of hyperthermophiles to adapt to their high-WO42environment.
Denitrification is an important anaerobic respiratory pathway employed by a diverse group of bacteria and archaea. The four enzymes, Nar, Nir, Nor, and Nos, catalyzing the entirety of this pathway reduce nitrate via nitrite, nitric oxide, and nitrous oxide to dinitrogen gas, respectively (1-5). The denitrification pathway and the properties of the four enzymes involved have been studied extensively in Gram-negative Proteobacteria, including Paracoccus denitrificans, Paracoccus halodenitrificans, Ralstonia eutropha, Pseudomonas aeruginosa, and Pseudomonas stutzeri, and are reviewed in refs (1-3). The pathway is, however, less understood in Gram-positive bacteria and archaea (2, 4-8). In contrast to bacterial denitrification enzymes, all four denitrification pathway enzymes in Pyrobaculum aerophilum directly couple to the quinone pool (4, 5, 7). With the exception of nitrate reductases and quinol:NO reductases, bacterial denitrification enzymes use periplasmic c-type cytochromes or blue copper proteins as electron donors (1-3). In bacteria, these electron donors are reduced by the electrogenic cytochrome bc1 complex, producing an overall
Hþ/e stoichiometry of 1 for the complete reduction of nitrate to dinitrogen gas from quinol (2). The first step of the bacterial denitrification pathway, catalyzed by the membrane-bound NarGHI-type nitrate reductase (Nar),1 is directly coupled to the generation of a proton motive force to sustain cell growth. The NarGHI-type Nar is largely conserved among all nitrate respiring bacteria and archaea (9-11). The enzyme has been extensively studied in mesophilic nitrate reducing bacteria such as the ammonifier Escherichia coli and the denitrifiers Pa. denitrificans, Ps. stutzeri, Pseudomonas denitrificans, and others (10, 11). The E. coli NarGHI enzyme has been recently crystallized and serves as the prototype in our understanding of the structure and function of nitrate reductases in other bacteria (12, 13). The quaternary structure of NarGHI is a dimer of heterotrimers. The largest subunit of the heterotrimer is the 139 kDa NarG, which contains a Mo-bis-MGD and a single [Fe-S] center (FS0). The 58 kDa NarH subunit harbors four distinct [Fe-S] centers, three [4Fe-4S] centers (FS1, FS2, and FS3), and one
This work was supported by grants from the National Science Foundation (MCB-0345037) to I.S. and The Netherlands Organization for Scientific Research (NWO 700.54.003) to S.d.V. *To whom correspondence should be addressed: Department of Microbiology, Immunology, and Molecular Genetics, UCLA, BSRB, RM 290, 615 Charles E. Young Dr. S., Los Angeles, CA 90095. Phone: (310) 206-0319. Fax: (310) 267-2774. E-mail:
[email protected].
1 Abbreviations: Nar, nitrate reductase; Nir, nitrite reductase; Nor, NO reductase; Nos, N2O reductase; BV, benzyl viologen; MV, methyl viologen; PB, plumbagin; W-Nar, tungsten-containing Nar; Mo-Nar, molybdenum-containing Nar; DMSO reductase, dimethyl sulfoxide reductase; TMAO reductase, trimethylamine N-oxide reductase; FDH, formate dehydrogenase; MQ, menaquinone; pmf, proton motive force; bis-MGD, bis-molybdopterin guanine dinucleotide; GMP, guanosine monophosphate.
r 2010 American Chemical Society
Published on Web 09/23/2010
†
pubs.acs.org/Biochemistry
9912
Biochemistry, Vol. 49, No. 45, 2010
[3Fe-4S] center (FS4). The 26 kDa NarI hydrophobic subunit accommodates a diheme b. NarI serves as the membrane anchor for NarGH and mediates the transfer of electrons from the quinol pool across the membrane to the [Fe-S] centers in NarH and subsequently to the Mo active site in NarG according to the QH2 f bD f bP f FS4 f FS3 f FS2 f FS1 f FS0 f Mo-bisMGD f NO3- pathway (12). To date, few archaeal Nars have been characterized. A threesubunit Mo-containing Nar was isolated from P. aerophilum (14), and two-subunit NarGH-type enzymes were purified from the halophilic archaea Haloarcula marismortui and Haloferax mediterranei (15-17). A distinctive feature of the H. marismortui and Ha. mediterranei nitrate reductases is their outward-facing active site (16, 18). In contrast, all bacterial nitrate reductases of the NarGHI type face the cell’s interior and catalyze nitrate reduction in the cytoplasm (11, 19). An additional interesting feature of archaeal Nars is the apparent absence of a narI-type gene encoding the membrane anchor subunit found in bacterial Nars and thus raises the question of how archaeal Nars are bound to the membrane. Recently, Yoshimatsu et al. purified a cytochrome b-561, NarC, from membranes of H. marismortui (20). On the basis of gene clustering of narC with narGH and homology to the bc1 complex cytochrome b, the authors suggested that in H. marismortui NarGH may form a respiratory supercomplex with the bc1 complex. Martinez-Espinosa et al. have proposed that two genes flanking narGH in Ha. mediterranei may serve to anchor the enzyme to the cytoplasmic membrane and directly accept electrons from an associated bc1 complex (18). Nitrate respiration at high temperatures presents a challenge; typical high-temperature environments are enriched with tungstate but depleted of molybdate (21). Because of its similar chemistry, WO42- usually acts antagonistically to MoO42-. In the presence of WO42-, bacteria cannot assemble an active Nar (22, 23). The hyperthermophile P. aerophilum is a denitrifying archaeon requiring WO42- for growth (6). Afshar et al. demonstrated that the external WO42- concentration affects the denitrification pathway efficiency of this archaeon, resulting in complete denitrification only at high WO42- concentrations. Nar purified from P. aerophilum grown with MoO42- and a limiting WO42- concentration (0.5 μM) was shown to contain Mo as a metal cofactor and a heme b membrane anchor (14). We report here that the Nar purified from P. aerophilum grown in the absence of added MoO42- and with 4.5 μM WO42- is a W enzyme. The W-Nar is identical to the Mo-Nar, indicating that either metal can serve as the active site ion. W is coordinated by bis-molybdopterin guanine dinucleotide (W-bis-MGD). The heme b was further characterized and identified as heme oP. Heme oP is coordinated by a membrane anchor, NarM, a homologue of one the proteins suggested to associate with the membrane of Ha. mediterranei Nar (18). The narM gene is conserved in all archaeal Nar operons identified thus far, and its gene product is related to electron transfer subunits of bacterial, periplasmic oxidoreductases. This is the first documentation of an active W-Nar and presents fascinating insight into how hyperthermophiles adapt to the metal composition of their environment. The coordination of tungsten as well as evolutionary and bioenergetic implications is discussed. EXPERIMENTAL PROCEDURES Growth Conditions. P. aerophilum (DSM 7523) was cultured in a marine medium at 95 C under anaerobic conditions as
de Vries et al. previously described (6) but with 4.5 μM Na2WO4 and the following modifications: NaBr, SrCl2 3 6H2O, NaF, KI, and Na2MO4 3 2H2O omitted from the medium. For the purification of the W/Mo-Nar, P. aerophilum was cultured with 1 μM Na2WO4 and 0.5 μM Na2MoO4. For large scale culture, P. aerophilum was grown in 70 L of medium in a 100 L glass-lined fermentor (Pfaudler). Growth was monitored by measurement of the optical density at 600 nm. Cells were harvested in the late exponential growth phase at an A600 of 0.28-0.32. After being harvested by concentration with a hollow-fiber filter (A/G Technology), cells were centrifuged for 40 min at 15000g and stored at -80 C. Purification of Nitrate Reductase. The enzyme purification procedure was performed under aerobic conditions and at 4 C unless indicated otherwise. All buffers contained 0.5 mM phenylmethanesulfonyl fluoride (PMSF). Initial purification steps, including HiTrapQ Sepharose chromatography, were performed essentially as previously described by Afshar et al. (14) with the following modifications. Cell lysis was enhanced by French press treatment, and membrane-bound protein was extracted using 3% Triton X-100. While the first chromatography step was essentially as described, enzyme purification was improved by employing a second HiTrapQ Sepharose (Pharmacia) chromatography step using 20 mM Tricine (pH 7.6) with 0.2% Triton X-100 as the buffer and a 0 to 0.3 M NaCl gradient, in which nitrate reductase eluted with ∼70 mM NaCl. Following a protein concentration step with Centricon-100 concentrators (Amicon), protein was separated on a Superdex 200 column (Pharmacia) using 20 mM Tricine (pH 7.6), 0.05% Triton X-100, and 0.15 M NaCl. Nitrate reductase fractions were combined, concentrated, and stored at 4 C and for the long term at -80 C. Analytical Protein Assays. Nitrate and nitrite reductase activities were assayed anaerobically at 75 C using reduced benzyl viologen (BV) or methyl viologen (MV) as the electron donor as previously described (14). In addition, nitrate reductase activity was assayed with reduced plumbagin as the electron acceptor (8). Succinate dehydrogenase activity was measured in the reverse direction with reduced benzyl viologen as the electron donor essentially as described elsewhere except that the assay was performed at 75 C (24). Malate dehydrogenase activity was measured at 75 C with NADþ as the electron acceptor as described previously (24). Formate dehydrogenase activity was measured anaerobically and at 75 C with methylene blue and dimethyl menaquinone (DMN) as described in ref 24. One unit of enzyme activity is defined as 1 mmol of substrate (nitrate, nitrite, fumarate, formate, or malate) reduced or oxidized per minute. The qualitative electron donor screen was performed anaerobically and at 80 C in capped serum vials with P. aerophilum membrane fractions. The assay contained 50 mM phosphate buffer (pH 7.0), 10 mM sodium nitrate, and the following sodium salts at 50 mM: formate, acetate, tartrate, DL-malate, pyruvate, DL-lactate, and succinate. In addition, NADH and NADPH at 10 mM and 0.1% casamino acids were tested as potential electron donors. After 15 min, the reaction was stopped when the vials were transferred to ice. The membrane fraction was removed by centrifugation, and nitrite formation was assessed with sulfanilamide and N-(1-naphthyl)ethylenediamine (24). The protein concentration was determined with the DC Bio-Rad protein assay (Bio-Rad) using bovine serum albumin (Sigma) as the standard. Iron, Mo, and W were quantified by inductively coupled plasma mass spectrometry at the Soil and Plant Analysis Laboratory of the University of Wisconsin/Extension (Madison, WI).
Article Pyridine hemochrome analysis was conducted as described previously (8). N-Terminal amino acid sequencing was performed with the protein obtained after the last purification step. Protein was incubated with SDS Laemmli buffer for 1.5 h at 90 C and separated by sodium dodecyl sulfate-polyacrylamide electrophoresis (SDS-PAGE) using 4 to 12% Bis-Tris gels (NuPAGE, Invitrogen). Protein was subsequently transferred to a polyvinylidene difluoride membrane (PVDF, Bio-Rad). Sequencing was conducted at the Microchemical Core Facility of the Keck School of Medicine at the University of Southern California/Norris Comprehensive Cancer Center. Guanine Identification and Characterization. To characterize the metal pteridin cofactor, guanine analysis was performed essentially as described by Hilton and Rajagopalan (25). To release the cofactor, purified nitrate reductase was hydrolyzed by addition of 0.1 M HCl and subsequent incubation at 100 C for 1 h. The sample was cooled to room temperature, and the pH was adjusted to 8.5 with 1 M NaOH. The centrifuged sample was applied to a 4.1 mm 300 mm Versapack 10 μm C-18 reversephase HPLC column (Alltech) equilibrated with 50 mM ammonium acetate and 7% methanol at a flow rate of 1 mL/min. The absorbance was monitored at 280 nm with a Shimadzu UV detector. Guanine was identified by retention time on the HPLC column and characteristic UV-vis spectrum compared to those of a standard (Sigma). For quantification, the integrated area of the absorbance peak associated with the eluting compound was calculated. It was compared to a calibration curve generated from guanine standards. Mass Spectrometry. Nar subunits were separated via SDSPAGE. Protein bands corresponding to NarG and NarM were excised, treated with trypsin, and analyzed by LC-MS/MS (liquid chromatography and tandem mass spectroscopy) essentially as described by Shevchenko et al. (26). For heme identification, hemes were extracted according to the procedure from L€ubben and Morand (27). Extracted hemes were analyzed by size exclusion chromatography and electrospray ionization mass spectrometry (SEC-MS) as described previously (28). A sizing column (Super SW 2000, 4.6 mm 300 mm, 40 C Tosoh Biosep) was equilibrated in a chloroform/ methanol/1% formic acid mixture in water (4:4:1, v/v/v) at a rate of 250 μL/min prior to injection of the extracted heme sample (100 μL). The orifice potential (65 V) used was typical of values employed for protein ionization. EPR Spectroscopy and Redox Titrations. EPR spectra were recorded with a Varian E-9 spectrometer at X-band frequency equipped with a home-built helium flow system (29). Sample reduction and reoxidation were conducted anaerobically via incubation for 1 min at 40 C with 1.0 mM plumbagin (PBH2) and 1.5 mM KNO3, respectively. The pH of purified Nar EPR samples was adjusted by addition of small aliquots of 0.5-1.0 M buffers to give a final concentration of 0.1 M (MES at pH 6, MOPS at pH 7, and Tris at pH 8). Membranes adjusted to various pH values were washed three times in one of these three buffers at 0.1 M. Dyemediated redox titrations (oxidative and reductive) monitored by EPR spectroscopy were performed in the additional presence of 0.1 M NaCl and with each of the following dyes at 80 μM: PES (phenazine ethosulfate), PMS (phenazine methosulfate), DCIP (2,6-dichlorophenolindophenol), methylene blue, resorufine, indigo carmine, 2-hydroxy-1,4-naphthaquinone, anthraquinone 2-sulfonate, phenosafranin, safranin, neutral red, benzyl viologen, and methyl viologen. Dye-mediated redox titrations monitored by
Biochemistry, Vol. 49, No. 45, 2010
9913
UV-vis spectroscopy were conducted in the presence of 20 μM PMS, PES, 2-hydroxy-1,4-naphthaquinone, anthraquinone 2-sulfonate, and 40 μM duroquinone. Midpoint potentials are quoted versus the normal hydrogen electrode (NHE) and were measured with a Pt electrode using saturated Ag/AgCl as a reference electrode; the redox potential was varied via addition of concentrated aliquots of sodium dithionite or ferricyanide. Dye-mediated redox titrations were conducted anaerobically in a Coy anaerobic chamber. The Nernst equation was used to calculate the midpoint potentials. Quoted midpoint potentials are (10 mV, except for FS3 ((20 mV). Simulation of EPR spectra was performed with a home-written program for the Macintosh computer operating under OS 9.2.2. RESULTS Purification of Nitrate Reductase from Cells Grown with 4.5 μM WO42- without Added MO42-. Unlike bacteria, P. aerophilum maintains an active nitrate reductase (Nar) even when grown at WO42- concentrations as high as 5 μM (6). Nar from P. aerophilum cultured with MO42- and under WO42concentration-limiting conditions (0.5 μM) was previously purified and shown to contain Mo in addition to Fe as metal cofactors (14). The purpose of this study was to examine the cofactor composition and kinetic properties of Nar from cells grown at high WO42- concentrations. P. aerophilum was grown anaerobically with nitrate in a medium containing 4.5 μM WO42but without added MoO42-. During initial purification attempts, we noticed that Nar from cells grown at high WO42- concentrations appeared to be more labile than the previously isolated MoNar; i.e., the enzyme gradually lost activity (14). Therefore, the purification procedure was modified to prevent enzyme inactivation. One of the modifications was the omission of hydroxyapatite chromatography known to delipidate membrane proteins, which resulted in an irreversible aggregation of Nar with concomitant loss of activity. Many tungsten-containing enzymes are inactivated by oxygen (21). We, therefore, attempted purification of Nar with N2-flushed buffers and in the presence of 5 mM dithiothreitol (DTT). However, DTT caused the complete loss of enzyme activity, and N2-flushed buffers did not increase Nar stability. Nar was purified by two anion exchange chromatography steps and one size exclusion chromatography step (Table S1 of the Supporting Information). After the final chromatography step, the enzyme was enriched 24-fold as compared to the membrane fraction with a purity of 80% approximated by SDS-PAGE (data not shown). Nar purity is consistent with the cofactor composition summarized in Table 1. Purified Nar has a subunit composition identical to that of the enzyme that was purified from cells grown at 0.5 μM WO42-, i.e., 130, 52, and 32 kDa (14). Subunit Identification and Operon Organization. The N-terminal sequence of the 52 kDa subunit was determined to be MNVRAQITMAMNLDK, which corresponds to the sequence predicted by the P. aerophilum narH gene. However, the N-terminal sequence of the 130 kDa subunit was blocked, and attempts to obtain the N-terminal sequence for the smallest subunit were unsuccessful. Therefore, the 130 and 32 kDa subunits were digested with trypsin within the SDS-polyacrylamide gel, extracted, and subjected to mass spectrometry analysis. Masses for 17 distinct peptides were obtained from the 130 kDa subunit, including YTDLPFLVILEPAGDGTYLQGR identifying NarG. The 29
9914
Biochemistry, Vol. 49, No. 45, 2010
de Vries et al.
Table 1: Cofactor Composition and Kinetic Parameters of P. aerophilum Nars cofactor
W-Nar (mol/mol)
Mo-Nara (mol/mol)
W Mo non-heme Fe GMP heme op
0.55
